Project Summary/Abstract Vaccines are a remarkably effective way to stem the threat posed by infectious diseases. Methods that allow rapid development of vaccines are vital. Synonymous recoding of viral genomes is a recently developed, general, and highly promising strategy for producing live attenuated vaccines. From an antigenic perspective, the method is ideal because it leaves the amino acid sequence of the viral proteins identical to the circulating pathogenic form. A number of viruses have been attenuated by recoding with non-preferred codons or codon pairs, and at least eight studies have shown protective immunization of mice. Despite its demonstrated success, there are fundamental gaps in our knowledge: 1) no effort has been made to compare alternative recoding strategies within the same virus in the same study; 2) several potential methods of synonymous recoding have not been tested at all; 3) the way in which attenuation is affected by the combination of multiple recoded genes is not known; and 4) most importantly, it is unresolved whether viruses attenuated by synonymous recoding are robust to evolutionary recovery. This proposal tackles these gaps through three Specific Aims. Aim 1: Identify methods of synonymous recoding and associated sequence features that can be used to generate viral genomes with a targeted level of attenuation. This aim includes developing empirical measures of individual codon and codon pair effects on translation rate to guide attenuation. It will also test metrics that have not previously been used for synonymous recoding. Aim 2: Extend models of adaptive evolution to determine if the attenuating effects, within and among genes and transcripts, combine in additive or non-additive ways. Aim 3: Determine if some strategies of attenuation are more robust to recovery than others. This aim will focus on viruses attenuated in multiple regions and by multiple methods, and also determine if some recovery pathways are broadly beneficial. The project takes advantage of a bacteriophage model system with well-developed tools for genome manipulation and methods for rapid experimental evolution relative to eukaryotic viral systems (i.e., a hundred generations per day at very large population sizes). Achieving these three aims will yield approaches that can be applied to other systems for designing viruses with targeted levels of attenuation that are robust to evolutionary recovery. This research is a critical step toward the long-term goal of achieving a general strategy for fighting infectious diseases by precision design of live vaccines that do not re-evolve virulence when used in humans.